122 ALASKAN GLACIER STUDIES mountain wall without intervening depression. Further, the ice was perceptibly higher than in 1905. We could not determine the exact amount of thickening, though it was certainly not less than 100 feet where the glacier passes out from its mountain valley, but the ice surface was not so high as it was in 1906. Banded Moraines on Expanded Bulb in 1909. The most noteworthy change in condition of the Variegated Glacier in 1909, as compared with 1905, was that the fine colored, banded moraines that in the former year covered the ice between the interior flat and the mountain valley were now gone. We could no longer trace the former succession. The purple and green moraines were entirely destroyed, and the other moraines were so disordered and displaced that we could no longer correlate them with the moraines of 1905; but ablation was at work covering the ice surface with new moraines of different kind and color (Fig. 8). We crossed this glacier from the north side near the mouth of the mountain valley to the interior flat. There was abundant evidence that the glacier had been profoundly broken, but it was, nevertheless, a matter of unceasing astonishment that in so short an interval ablation had so effectually healed the broken glacier surface. Where the moraine cover was not thick enough to protect the ice, its surface was thrown into a series of great waves, 50 to 75 feet high, with the areas of greater crevassing mainly in the depressions. In addition, there were many smaller ice ridges and hummocks, all much rounded by ablation. The area of clear ice extended far down below its former position, apparently a mile or more, and the pronounced red moraine that formerly rose as the lower border of the clear ice had apparently been carried bodily forward and incorporated in the ice. A small area along the northern margin, which was not profoundly affected by the advance, was still stagnant and retained a series of banded moraines; but everywhere else inside the interior flat the old series of variegated moraines, from which the name of the glacier was derived, had been destroyed. In their place an entirely new set of wholly different character and arrangement had begun to develop. On the southeast side a band of white granite moraine appeared, but it did not swing out into the ice in crescenticform. Then, after an intervening area of clear ice (PL LVI), a band of black hornblende moraine swept in a great crescent far across the ice bulb, which was here expanded beyond the mountain front. This morainic crescent was not complete, being most pronounced toward the sides and dying out completely near the center, where in its place was a red gneiss moraine much thinner and less pronounced than the one of black hornblende. Both on its inner and outer face the black hornblende moraine rose about 100 feet above the clear ice. This gives a rough measure of the extent of ablation on this ice surface since 1906, for it is not to be supposed that the moraine had any notable initial elevation during the crevassing and forward motion of the ice. Although it is true that the ice under the ridge has also been somewhat lowered, it has been much protected by the debris, and in the meantime the clear ice has been lowered over 100 feet. Therefore, since 1906, the general ablation in this section has been in excess of 100 feet, and this fact gives a basis for understanding the remarkable effectiveness of ablation in smoothing the broken surface. One very noteworthy fact was the shallowness of the crevasses, for in 1909 the bottoms of even the deepest were visible. It was evidently only the rigid upper ice that was broken, and the breaking, though extending down over 100 feet, probably did not much exceed 200 feet.